Harvard Pilgrim researchers develop model to reduce errors from sound-alike, look-alike medications

A new tool could be on the horizon for identifying drugs that have similar names in order to help cut down on errors caused when prescription drugs are mixed up—or even preventing drugs from having names that are too similar to begin with.

Researchers at the Harvard Pilgrim Health Care Institute developed and validated a data-driven prediction model to accurately predict sound-alike/look-alike (SALA) medication pairs to proactively identify and prevent potential errors. Their work was presented at the American Society of Health-System Pharmacists (ASHP) summer meeting Monday.

“Even well-trained healthcare professionals can mistake one medication for another with an almost identical name," said Qoua Her, research analyst at the institute and a Harvard Medical School affiliate, in a statement. “Our new prediction model has the potential to reduce critical errors by identifying high-risk sound-alike-look-alike medication pairs.”

Mistakes from SALA medications contribute to as many as 250,000 hospitalizations each year from medication errors, according to the ASHP. Notably, in 2016, there were 55 reports of confusion between Brintellix, an antidepressant, and Brilinta, a blood thinner, including two serious adverse events. The U.S. Food and Drug Administration (FDA) approved a name change for Brintellix to Trintellix.   

RELATED: Pronovost's new mission: Convincing health systems to tell a different story

To conduct the study, the team from Harvard Pilgrim evaluated 82 medication name similarities and seven product attribute measures with 40,000 samples of medication pairs. They used that to create a prediction model comprised of 13 of the strongest predictors for potential medication errors, such as having the same first letter in the medication names or coming from the same manufacturer.